Merrimack River Watershed Council, Inc.

Transcription

1 Merrimack River Watershed Council, Inc. The Voice of the Merrimack Merrimack River Monitoring Program Formerly known as the Merrimack River Water Quality Monitoring, Analyzing, Protecting and Promoting (MAPP) Program 2009 Annual Report Prepared by: Merrimack River Watershed Council, Inc. 600 Suffolk Street, 5 th Floor Lowell, MA April 22, 2010

2 Acknowledgements The Merrimack River Watershed Council Inc. (MRWC) would like to thank those members of the Volunteer Environmental Monitoring Network (VEMN) who have helped us collect the water quality data described in this report, including the Lowell National Historical Park. Without the dedication of these volunteers, the Merrimack River Monitoring Program, formerly known as the Merrimack River Water Quality Monitoring, Analyzing, Protecting and Promoting (MAPP) Program, would not be possible. VEMN volunteers have donated time, boats, fuel, boat ramp fees, equipment, expertise and knowledge to make this program a success. MRWC thanks you on behalf of the Merrimack River and its communities, both human and natural, for your hard work. We would also like to acknowledge the support we have received from our funders. Financial donations from Massachusetts Environmental Trust, the Stevens Foundation, the Davis Conservation Foundation, the EnTrust Fund and the Cabot Family Charitable Trust have provided the resources necessary for the continued success of the Merrimack Monitoring Program in its third year. The Environmental Protection Agency (EPA) provided the Global Positioning System units needed for this project through its equipment loan program as well as donating laboratory bacteria analysis, ammonia test strips and a surfactant test kit as an in kind contribution. Finally, the Lowell Wastewater Treatment Utility and Waters Corporation have provided additional in kind support for nutrient, bacteria, trace metals and pharmaceutical sample analysis. Tracie Sales Water Resources Manager Christine Tabak Executive Director Merrimack River Watershed Council, Inc.

3 Table of Contents Merrimack River Monitoring Program Annual Report...1 Executive Summary...6 Introduction...8 Characteristics of the Merrimack River...8 Baseline Monitoring Project Project Location...10 Methods...13 Quality Assurance/Quality Control Baseline Water Quality Results and Discussion...17 Bacteria...17 Field Study Analysis...17 CSO Analysis...23 Temperature...26 Dissolved Oxygen...28 ph...29 Specific Conductance, Total Dissolved Solids & Salinity...32 Clarity...33 Continuous Water Quality Monitoring Hotspot Monitoring...36 Nutrients and Metals Screening Methods...38 Results and Discussion...39 Ammonia...39 Nitrogen...40 Phosphorus...40 Detergents...40 Metals...40 Pharmaceutical Product Screening Methods and Project Location...42 Pharmaceutical Screening Results and Discussion...43 Accomplishments and Next Steps Accomplishments...45 Next Steps...47 References Appendix A: Summary of Bacteria Results Appendix B: Water Quality Tables

4 List of Tables Table 1. List of 2009 Baseline monitoring stations in the Merrimack River Table 2. List of physical water quality parameters measured Table 3. List of bacteria sampling dates by section. Bold face indicates wet weather events Table 4. Massachusetts (MA DEP 2007) and New Hampshire (NH 1998) water quality standards Table 5. Summary of 2009 Merrimack River Baseline Water Quality Monitoring Project bacteria water quality results percent of time the Merrimack River meets water quality standards under various criteria Table 6. National and Massachusetts water quality criteria. Values are maximum allowable concentrations unless otherwise noted

5 List of Figures Figure 1. Map of 2009 Baseline monitoring stations on the Merrimack River Figure 2. Daily precipitation (bar graph) and average daily stream flow (line graph) in the Merrimack River May through October, 2009 in Lowell, Massachusetts (USGS Undated, NOAA Undated) Figure 3. Enterococcus bacteria concentrations May through October 2009 in the Merrimack River in Section 1 between Haverhill and Newburyport, Massachusetts. Hatch pattern indicates wet weather event. Values greater than 104 cfu/100ml indicate unsafe swimming conditions Figure 4. E. coli bacteria concentrations in the Merrimack River between May and October 2009 in Section 2 (Haverhill to Lawrence), Section 3 (Lawrence to Lowell), Section 4 (Lowell to Tyngsborough), and Section 5 (Nashua/Hudson). See Table 3 for wet versus dry weather events. Values over the red line indicate unsafe swimming conditions according to respective state water quality standards Figure 5. Geometric mean of E. coli and Enterococcus bacteria concentrations May through October 2009 in the Merrimack River. Stations with less than five samples, including all those in Section 5, were not included in the geometric mean calculation Figure 6. Total daily precipitation in inches in Lowell, Massachusetts compared to days on which LRWWU CSO diversions occurred May through October, Figure 7. Total daily precipitation in inches in Lawrence, Massachusetts compared to days on which GLSD CSO occurred May through October, Figure 8. Total daily precipitation in inches in Haverhill, Massachusetts compared to days on which HWTF CSO occurred May through October, Figure 9. Median daily water temperature in the Merrimack River between May and October 2009 combined with the average of the mean daily air temperature for the lower Merrimack Valley (Nashua, NH and Lowell, Lawrence, Haverhill and Groveland, MA) Figure 10. Median dissolved oxygen results by station in the Merrimack River between May and October Figure 11. Median ph at each monitoring station in the Merrimack River between May and October Figure 12. Median specific conductance for fresh water stations in the Merrimack River between May and October Figure 13. Percent dissolved oxygen saturation and daily precipitation in the Merrimack River s Pawtucket Dam impoundment in Lowell, Massachusetts from September 22 through October 6, Figure 14. Location and range of bacteria counts for several Merrimack River hotspot samples in

6 Figure 15. Aluminum levels in Section 4 (Tyngsborough Lowell) of the Merrimack River in Massachusetts on October 20, Red line indicates the EPA chronic exposure limit for aluminum in surface waters Figure 16. Location of pharmaceutical sampling sites on the Merrimack River, June 24, Figure 17. Pharmaceutical products found in Section 2 of the Merrimack River, June 24, Figure 18. Number of pharmaceutical products found at each site Section 2 of the Merrimack River, June 24, List of Water Quality Results Tables Bacteria Results Water Temperature Results Dissolved Oxygen Results ph Results Specific Conductance Results Total Dissolved Solids Results Salinity Results Nutrient Results Metals Results

7 Executive Summary The Merrimack River Monitoring Program is a volunteer water quality monitoring effort begun in 2007 to collect baseline water quality information in the 50 mile mainstem of the Merrimack River in Massachusetts. Since its inception the program has expanded geographically to include monitoring in southern New Hampshire and programmatically to incorporate additional water quality parameters. In 2009 alone, 40 Merrimack Valley community members volunteered with the Merrimack River Watershed Council (MRWC) to collect water quality data at 41 sites along the length of the river. Volunteer teams monitored seven to nine sites in one of five river sections from Newburyport to Nashua, traveling from one site to another via boat. Over forty monitoring trips occurred throughout the spring, summer and fall of 2009, with bacteria samples collected on 23 of these trips, nutrient data collected on five days, pharmaceutical product samples collected once and physical water quality parameters recorded on all of the days. Physical water quality data collected includes water temperature, ph, dissolved oxygen, conductivity, total dissolved solids, salinity and Secchi depth. Physical water quality parameters met state standards with the exception of ph. On several days between May and August ph values as low as 3.2 to 4.1, the acidity of vinegar, were found in various parts of the river. Bacteria samples were collected once per month in each section and analyzed at the Region 1 EPA laboratory. In comparison to the data MRWC collected in prior years, 2009 dry weather bacteria results remained relatively consistent for the number of days the river was safe for swimming (96 percent) and boating (99 percent) according to state water quality standards where the sample was collected. In wet weather, 2009 data indicated an improvement in Merrimack River water quality: the river met state water quality standards 95 percent of the time for swimming and 100 percent of the time for boating. Evaluation of Merrimack River water quality based solely on criteria used by New Hampshire or by other Massachusetts watershed associations would indicate lower water quality, however, with only two thirds of wet weather days safe for swimming. In 2009 MRWC also had the exciting opportunity to get its new Safe Drinking Water Project off to a flying start with a screening for pharmaceutical products. Samples collected at several locations between Lawrence and Haverhill, Massachusetts, an area downstream of several drinking water sources, came back positive for 16 of 20 common drugs. A few of our 2009 discoveries and successes include: Septic leak in Lawrence fixed MRWC identified a pipe discharging polluted effluent into the Merrimack River in Lawrence. By working with Massachusetts Department of Environmental Protection officials in cooperation with the City of Lawrence, the source of the leaking septic system was determined and the leak fixed. 6

8 Spicket and Shawsheen Rivers contribute pollution to the Merrimack 2009 bacteria data, supported by a geometric mean of results over the Massachusetts state water quality limit, suggests that both the Spicket and Shawsheen Rivers frequently contain high levels of bacteria. Both rivers also demonstrate significantly higher conductivity and total dissolved solids than the Merrimack mainstem. Sampling along the length the tributaries will be necessary to pinpoint specific sources. Merrimack River nutrient and metals monitoring begun Nutrient and metals monitoring began in 2009 as part of the Search and Restore Project. Results collected in 2009 have provided baseline nutrient data for the river and identified critical stations to target for the wet and dry weather monitoring planned for Analysis of metals in the water has also identified aluminum as a potential element of concern. The first three years of the Merrimack River Monitoring program have reestablished MRWC s Volunteer Environmental Monitoring Network and effectively engaged local community organizations and citizens regarding water quality concerns in the river. Future plans include continuing baseline monitoring and bacteria sampling in Massachusetts and southern New Hampshire, nutrient monitoring and intensive sampling of high use and problem sites on the river. MRWC will also continue spreading information about the work that yet needs to be done to achieve our vision of a pure Merrimack River, respected and enjoyed. 7

9 Introduction The Merrimack River Watershed Council, Inc. (MRWC) is a non profit 501(c)(3) organization formed in 1976 by local activists and regional planning commissions to promote citizen involvement in the clean up of the Merrimack River. Its organizational mission today is to ensure the sustainable ecological integrity and balanced, managed use of the Merrimack River and its watershed through science, advocacy, partnering and recreation. Our focus area is the Merrimack River Watershed mainstem and its adjoining communities in Massachusetts and New Hampshire, though we have also accomplished many projects in our eighteen sub watersheds. We understand that we are the only thirdparty advocate of the entire length of the Merrimack River in Massachusetts who is independent of commercial or regulatory interests; we are The Voice of the Merrimack. Since the mission of the MRWC is to ensure the integrity and balanced use of the watershed and its resources, it is imperative that we focus on the river from which our organization is named. In the past, MRWC has performed extensive projects on tributaries of the Merrimack River, while leaving the health of the Merrimack River itself relatively unchecked. In 2007, the board and staff chose to rectify this past oversight by committing to the Merrimack River Water Quality Monitoring, Analyzing, Protecting and Promoting (MAPP) Project, now renamed the Merrimack River Baseline Monitoring Project. The Baseline Monitoring Project is a three phase program designed to: (1) quantify the baseline water quality of the Merrimack River, (2) discover sources of pollution to the river, address and reduce pollution to the Merrimack River through both traditional and creative methods, and (3) educate watershed constituents on how to protect this important resource. Since 2007 MRWC s water quality monitoring efforts have grown to become the Merrimack River Monitoring Program, encompassing the original Baseline Monitoring (MAPP) Project, the Merrimack River Search and Restore Project (formerly known as the Impairment Quantification or IQ Project), and the Safe Drinking Water Project. The main body of this report summarizes the results of the 2009 Baseline Monitoring Project, which includes recent expansion of baseline water quality monitoring into southern New Hampshire. Additional sections review the initial nutrient and metal sampling results of the first year of the two year Search and Restore Project as well as the results of the pharmaceutical screening conducted in June 2009 as the first phase of the Safe Drinking Water Project. Characteristics of the Merrimack River The Merrimack River is 115 miles long, beginning at the confluence of the Pemigewasset and Winnipesaukee Rivers in Franklin, New Hampshire and flowing approximately 65 miles in New Hampshire and another 50 miles in Massachusetts to its mouth in Newburyport, Massachusetts. There are a total of six dams on the mainstem of the river, though only two in the stretch of river monitored by MRWC: the Essex Dam in Lawrence, Massachusetts and the Pawtucket Dam in Lowell, Massachusetts. There are two USGS gauging stations on the Merrimack River in Massachusetts, one in downtown Haverhill that measures only water height due to the influence of the tides, and one in Lowell at the confluence with the Concord River that measures stream flow. A new 8

10 gauging station was installed during the summer of 2009 on the Merrimack in downtown Nashua, New Hampshire. Between Newburyport and the Essex Dam in Lawrence, the river is affected by ocean tides. Salt water intrudes up the river five to ten miles depending on the tide and river volume, and the river current can reverse, depending on the height of the tide and the level of flow in the river, up to the Mitchell s Falls area in Haverhill when the tide comes in. Water levels in the river can be tidally affected for the entire 29 miles from the estuary in Newburyport up to the Essex Dam during periods of low flow, such as during a typical August and September (M. Vets, Haverhill Harbormaster, anecdotal). In general, high tide in Haverhill lags high tide in Newburyport by approximately 1¼ hours, while low tide in Haverhill lags low tide in Newburyport by approximately 3 hours. 9

11 Baseline Monitoring Project Project Location The 2009 Baseline Monitoring Project collected water quality information in the mainstem of the Merrimack River in Massachusetts and southern New Hampshire. Figure 1 illustrates the 41 monitoring stations, each near an outfall, tributary or at a historical monitoring site. Monitoring occurred regularly between May and October at most of the identified sites, though data was only collected in Section 2 between June and October. Two stations in Massachusetts, 38.9 and 40.0, and five stations in New Hampshire, 51.8 through 55.9, were only monitored once due to access and boat availability difficulties. Monitoring was conducted in five river sections, with 7 to 9 sites located in each section. The river sections are: 1) the estuary in Newburyport to the Haverhill/Groveland town line, 2) Haverhill to the Essex Dam in Lawrence, 3) the Essex Dam to the Pawtucket Dam in Lowell, 4) the Pawtucket Dam to the Massachusetts/New Hampshire state border, and 5) the state border to Greeley Park in Nashua. Stations in section 5 were monitored for the first time in 2009 as the program was expanded to encompass the Nashua and Hudson area of southern New Hampshire. Two new stations were also added in section 3 in Lowell between the Pawtucket Dam and Duck Island. Because of shallow water, this area is usually inaccessible via motor boat, but can be reached by paddlers. In section 4, the station located at the Lowell water intake, only 0.2 miles upstream of the Stony Brook station, was removed and replaced with station 44.6 at the Vesper Country Club. In Section 2, station 27.8 was added at the mouth of the Shawsheen River as a result of high levels of bacteria found in hotspot (areas of known or suspected high pollution) samples collected in the tributary. Finally, station 3.8 in Newburyport was eliminated for its proximity to station 4.4 and a new station called Kimball Farm was created upstream at mile 11.8 near Rocks Village, an area that was not being tested. Table 1 lists the stations monitored in

12 Figure 1. Map of 2009 Baseline monitoring stations on the Merrimack River.

13 Table 1. List of 2009 Baseline monitoring stations in the Merrimack River. Section Station Description Town 2.7 Newburyport Waste Water Treatment Plant Newburyport 4.4 Yankee Marina Newburyport 6.8 Powow River Amesbury Artichoke River Newburyport 9.4 Indian River West Newbury 10.6 Cobbler Brook Merrimac 11.8 Kimball Farm Merrimac 14.1 Old North Canal Haverhill 16.8 Johnson Creek Groveland 17.8 Haverhill Waste Water Treatment Plant Haverhill 19.1 Little River Haverhill Creek Brook Haverhill 25.6 Lucent Technologies North Andover 26.9 Greater Lawrence Wastewater Treatment Plant North Andover 27.8 Shawsheen River Lawrence 28.2 Spickett River Lawrence 29.1 Below Essex Dam Lawrence 29.6 Above Essex Dam Lawrence 31.4 Methuen Water Intake Methuen 32.2 Bartlett Brook Methuen 33.4 Fish Brook Andover Gravel Pit Dracut 36.3 Trull Brook Tewksbury 37.9 Duck Island Lowell 38.9 Concord River Lowell 40.0 Oulette Bridge Lowell 41.1 Pawtucket Dam Lowell 42.4 Rourke Bridge Lowell 43.4 Stony Brook Chelmsford Vesper Country Club Lowell 46.4 Lawrence Brook Tyngsborough 47.3 Tyngsborough (Rte. 113) Bridge Tyngsborough 48.9 Limit Brook Tyngsborough 49.6 Massachusetts/New Hampshire Border Tyngsborough 49.9 Pheasant Lane Mall Nashua 50.9 Spit Brook Nashua Unnamed Stream Hudson 52.5 Nashua Country Club Nashua 53.1 Nashua WWTP Nashua 54.4 Nashua River Nashua 55.9 Greeley Park Nashua 12

14 Methods All water quality monitoring and sampling was conducted in accordance with the procedures outlined in the Massachusetts Department of Environmental Protection (MA DEP) and the U.S. Environmental Protection Agency (EPA) approved 2007 Quality Assurance Project Plan (QAPP) for the Baseline Monitoring/MAPP Project, though data from two monitoring stations were collected via a non motorized vessel. Details of the data collection methods and analysis can be found in the MRWC QAPP, available upon request from MRWC. Physical water quality parameters outlined in Table 2 were collected in situ with YSI 556 water quality probes and secchi disks. Measurements of all physical water quality parameters except for clarity were made beginning at the surface and extending down as far as possible without touching the bottom in one meter intervals to three meters. After the three meter measurement, the interval increased to every other meter (i.e. 5m, 7m and 9m), though in some cases a four meter reading was collected when the depth of the river was between four and five meters. In most cases, the median value of all of the depths measured at each station on each day has been reported wherever the percent difference between the median and the individual values was less than 10%. Where the percent difference between the median and individual values was greater than 10%, such as in the tidally affected portion of the river, depth data was maintained for analysis. Median data for 2009 can be found in Appendix A: Water Quality Tables. Table 2. List of physical water quality parameters measured. Parameter Equipment Water Temperature YSI 556 Dissolved Oxygen YSI 556 Specific Conductivity YSI 556 Total Dissolved Solids YSI 556 Salinity YSI 556 ph YSI 556 Clarity Secchi Disk Monitoring occurred on a total of 32 days in On seven of these days, monitoring occurred simultaneously in two or three sections: Section 1 14 days Section 2 5 days Section 3 7 days Section 4 8 days Section 5 5 days Of the 40 monitoring trips on the Merrimack River in 2009, 20 occurred on dry weather days, defined as less than 0.25 inches of precipitation falling in the 72 hours 13

15 prior to the monitoring day, and 20 occurred after wet weather events. Precipitation amounts were calculated for each river section based on triangulation principals using the daily totals recorded at the nearest two or three weather stations (NOAA Undated). Stations used were located in Newburyport, Groveland, Haverhill, Lawrence and Lowell, Massachusetts as well as in Nashua, New Hampshire. Figure 2 shows daily rainfall (inches) and river flow rate (cubic feet per second) in Lowell from May 1, 2009 through October 31, Bacteria grab samples were collected, generally once per month, just below the surface of the water at each station and were analyzed at the EPA Region 1 laboratory in North Chelmsford, Massachusetts. Table 3 lists the dates on which samples were collected in each section. Samples were analyzed for Escherichia coli (E. coli) in the fresh water portions of the Merrimack River from Haverhill to Nashua (sections 2 through 5) and Enterococcus in the salt water section of the river from Newburyport to Haverhill (section 1). Of the 21 days (including 27 sampling events) on which bacteria samples were collected, 11 days were considered dry weather and 10 days were considered wet weather. Wet weather dates are indicated in boldface in Table 3. Table 3. List of bacteria sampling dates by section. Bold face indicates wet weather events. Section May June July August September October 1 5/20, 5/27 6/24 7/22 8/26 9/23 10/28 2 n/a 6/24 7/23 8/19 9/16 10/27 3 5/14 6/11 7/23 8/13 n/a 10/19 4 5/12 6/10 7/14 8/11 9/8 10/20 5 5/12 n/a n/a 8/11 9/8 10/20 Quality Assurance/Quality Control Quality assurance was achieved by collecting duplicate samples and readings as well as through the oversight of the Baseline Monitoring Project quality control officer. The MRWC QAPP requires that 10% of the physical water quality data collected with the YSI 556 units be duplicate measurements, and in 2009 volunteers collected 20% duplicate samples. Of the 1320 duplicate measurements collected and validated, all but 23, or less than two percent, were within the precision set for acceptable variability (10% for water temperature, 20% for the other parameters). Of these 23 measurements, 19 were collected in the Newburyport to Amesbury stretch of river where the influence of ocean water is greatest and a slight disturbance in the water as the probe passes through salt lens can alter the results dramatically. The data from these sampling points for the affected parameters (primarily salinity, specific conductivity and total dissolved solids) were not included in the analysis. Three other duplicate measurement exceptions were caused by operator error when probes lowered into the water for the first time on the trip were not given enough time for the readings to settle before the results were recorded. These measurements were removed from the data analysis as incorrectly collected data and the volunteers re trained in the necessary collection techniques. The final excessive 14

16 0 5/1 5/13 5/25 6/6 6/18 6/30 7/12 7/24 8/5 8/17 8/29 9/10 9/22 10/4 10/16 10/28 45,000 40, ,000 Precipitation (Inches) 2 3 Precipitation Discharge 30,000 25,000 20,000 15,000 Daily Mean Flow (Cubic Feet/Sec.) 4 10,000 5,000 5 Figure 2. Daily precipitation (bar graph) and average daily stream flow (line graph) in the Merrimack River May through October, 2009 in Lowell, Massachusetts (USGS Undated, NOAA Undated). 0

17 variance in the duplicate data occurred for a salinity reading taken in fresh water where the value was at the extreme low end of the equipment s detection limit. The 0.01 ppt difference between the original and duplicate values has been deemed insignificant despite the mathematical calculation that causes the difference to appear significant. Quality control for water clarity was achieved by having two volunteers observe the depth of the Secchi disk on 100% of the readings. Because Secchi depth readings can vary significantly depending on whether the readings were taken in sunshine versus shade, on smooth water versus choppy water, or in calm water versus a strong current, Secchi depth data analysis is not being included in this report except on an anecdotal basis. While attempts are made to make measurements as consistent as possible, MRWC has determined that the Secchi disks are not only difficult to use but the results are also inconclusive in a river with a strong current. Volunteers also collected a duplicate sample for bacteria analysis on each sampling trip at one of the monitoring stations visited, with 11% of all samples being duplicates. For all but five duplicate bacteria samples, results fell within of the acceptable precision range of 30% for log transformed data. For those five duplicates outside of the acceptable range, all concentrations were within the water quality limits set for safe swimming, and four of the five were very low values. With very low values, the relative percent difference technique used for quality assurance/quality control comparison is less effective because a small difference in values becomes a large percent difference. Because these four samples contained small bacteria concentrations only slightly above the minimum detection limits [the least number of colonies that can be counted: four colony forming units per 100 milliliters of water (cfu/100ml) for E. coli and ten cfu/100ml for Enterococcus], MRWC assumes that this data is valid. The fifth duplicate sample outside of acceptable precision was an Enterococcus sample collected at station 2.7 near the outfall of the Newburyport wastewater treatment plant. Given reports of a strong current at the sampling site on that day and the location near the treatment plant outfall, the variation in the primary and duplicate samples may be due to natural variation in an area where mixing would be incomplete. For analytical purposes, the concentrations of all of the primary and duplicate samples have been averaged. One duplicate E. coli sample collected in Section 3 was not listed on the data sheets and thus there was no record of the site at which it was collected. All bacteria samples collected on that day had concentrations well below the state water quality standards, however, and at only one station would the difference between the original sample and the duplicate have exceeded the acceptable precision range. Therefore, bacteria data collected on this day is still considered valid. Data collection and data entry procedures were overseen by the Baseline Monitoring Project quality control officer. Data collection methods were deemed acceptable, and any errors in data entry were corrected. More details about the quality assurance practices for the project can be found in the MRWC QAPP, available upon request from MRWC. 16

18 2009 Baseline Water Quality Results and Discussion The Merrimack River is designated as a Class B (freshwater) warm water fishery in New Hampshire and in Massachusetts from the New Hampshire state border to Haverhill and a Class SB (tidally affected) water body from Haverhill to the estuary in Newburyport. This means that the river is expected to support fish, aquatic life and other wildlife as well as be suitable for primary (swimming) and secondary (boating) contact. Class B waters should also be suitable as a drinking water supply with adequate treatment, while Class SB waters should support conditional shellfish harvesting (MA DEP 2007). For this type of water body, each state has set limits for the amount of bacteria the water can safely contain, the maximum water temperature, the amount of dissolved oxygen in the water and the ph. These limits are listed in Table 4. Table 4. Massachusetts (MA DEP 2007) and New Hampshire (NH 1998) water quality standards. Parameter MA Limit NH Limit E. coli (fresh water bacteria, cfu/100 ml) 235 (swim) single sample 126 (swim) geometric mean 88 (swim) single sample 47 (swim) geometric mean 1260 (boat)* 10% samples 630 (boat)* geometric mean 406 single sample 126 geometric mean 104 (swim) single sample Enterococcus 35 (swim) geometric mean (salt water bacteria, cfu/100 ml) 350 (boat)* 10% samples 175 (boat)* geometric mean N/A Water Temperature < 28.3 C Class B warm < 29.4 C Class SB Supportive of Class B uses Dissolved Oxygen > 5.0 mg/l > 75% saturation > 5.0 mg/l during CSOs ph 6.5 < ph < 8.3 Class B 6.5 < ph < 8.0 Class B 6.5 < ph < 8.5 Class SB N/A * Bacteria safety limits for secondary contact/boating are based on Massachusetts Class C waters. Bacteria FIELD STUDY ANALYSIS Measurements of Escherichia coli (E. coli) and Enterococcus bacteria are used by the states of Massachusetts and New Hampshire to determine human health risks from primary (swimming) and secondary (boating) contact in fresh and salt waters, with E. coli used in fresh water and Enterococcus used in salt water. Both E. coli and Enterococcus are bacterium commonly found in the waste of warm blooded animals. While these strains of bacteria have not been identified as directly causing adverse health effects, they do indicate that other, more harmful, strains of bacteria are likely present. The states use two different standards to evaluate bacterial water quality, and also use different standards depending on the number of samples collected at the site. For class B (fresh) waters in Massachusetts the geometric mean of all E. coli samples taken within the most 17

19 recent six months shall not exceed 126 colonies per 100 ml typically based on a minimum of five samples and no single sample shall exceed 235 colonies per 100 ml (MA DEP 2007). For Massachusetts class SB (salt) waters no single Enterococci sample taken during the bathing season shall exceed 104 colonies per 100ml and the geometric mean of the five most recent samples taken within the same bathing season shall not exceed 35 Enterococci colonies per 100ml (MA DEP 2007). New Hampshire standards are more strict for fresh water where designated beach areas shall contain not more than a geometric mean based on at least 3 samples obtained over a 60 day period of 47 Escherichia coli per 100 milliliters, or 88 Escherichia coli per 100 milliliters in any one sample (NH 1998), but are the same for salt water, though the Merrimack watershed is entirely fresh water in New Hampshire. Of the samples MRWC collected during dry weather in 2009, the Merrimack River met single sample bacteria water quality standards for swimming 96 percent of the time and 99 percent of the time for boating. According to the samples MRWC gathered during wet weather, the river met single sample water quality standards 95 percent of the time for swimming and 100 percent of the time for boating. Figures 3 and 4 illustrate the 2009 single sample bacteria counts at each station for Enterococcus in Section 1 and for E. coli in Sections 2, 3, 4 and 5, respectively. In these calculations, Massachusetts water quality standards were used for those samples collected in Massachusetts while New Hampshire standards were used for samples gathered in that state. Table 5 summarizes the 2009 bacteria water quality results under these state standards as well as under the two more protective standards described below. Table 5. Summary of 2009 Merrimack River Baseline Water Quality Monitoring Project bacteria water quality results percent of time the Merrimack River meets water quality standards under various criteria. State Single Sample NH Standards CRWA Standards Weather Swim Boat Swim Boat Swim Boat Dry 96% 99% 85% 99% 74% 99% Wet 95% 100% 68% 98% 61% 98% If New Hampshire s water quality standards were used for bacteria results in both Massachusetts and New Hampshire, water quality in the Merrimack would appear to be lower. Using the E. coli 88 cfu/100ml standard for swimming and 406 cfu/100ml standard for boating in both states, while maintaining the Enterococcus 104 cfu/100ml (swimming) and 350 cfu/100ml (boating) standards, the 2009 data MRWC collected during dry weather indicates that the Merrimack River met bacteria water quality standards for swimming only 85 percent of the time but still met standards for boating 99 percent of the time. Similarly, MRWC wet weather samples under the New Hampshire criteria suggest that the river met water quality standards just 68 percent of the time for swimming but 98 percent of the time for boating. 18

20 200 May June 175 July August September Enterococcus Concentration (cfu/100ml) October MA State Limit for Swimming = Haverhill Station Newburyport Figure 3. Enterococcus bacteria concentrations May through October 2009 in the Merrimack River in Section 1 between Haverhill and Newburyport, Massachusetts. Hatch pattern indicates wet weather event. Values greater than 104 cfu/100ml indicate unsafe swimming conditions.

21 May June July August September October E. coli Concentration (cfu/100ml) State Limits for Swimming: NH = 88 MA = Nashua Lowell Station Lawrence Haverhill Figure 4. E. coli bacteria concentrations in the Merrimack River between May and October 2009 in Section 2 (Haverhill to Lawrence), Section 3 (Lawrence to Lowell), Section 4 (Lowell to Tyngsborough), and Section 5 (Nashua/Hudson). See Table 3 for wet versus dry weather events. Values over the red line indicate unsafe swimming conditions according to respective state water quality standards.

22 The Charles River Watershed Association (CRWA) uses the Massachusetts geometric mean bacteria limits, 126 cfu/100ml for E. coli, on single sample bacteria results rather than the single sample criteria, 235 cfu/100ml for E. coli, to determine whether or not the Charles River is safe for swimming or boating, arguing that these lower limits are more protective of human and river health (CRWA 2009). Under these criteria, assuming a similar use of the geometric mean limit of 35 cfu/100ml for single Enterococcus samples, the 2009 data MRWC collected during dry weather indicates that the Merrimack River met bacteria water quality standards for swimming only 74 percent of the time but still met standards for boating 99 percent of the time. Similarly, MRWC wet weather samples under the CRWA criteria suggest that the river met water quality standards just 61 percent of the time for swimming but 98 percent of the time for boating. Because MRWC was able to collect at least five samples at most Massachusetts stations in 2009, we were also able to calculate the geometric mean of bacteria counts for each station. Based on the 2009 geometric mean calculations, water quality at six stations exceeded Massachusetts state standards. As shown in Figure 5, three of these stations are located in Section 2, all of them at the mouth a major tributary (Spicket, Shawsheen and Little Rivers). Both the Spicket and Shawsheen Rivers have demonstrated water quality problems in the past and need to be monitored more intensely to track pollution sources within them. The Little River has not traditionally shown significant water quality problems, and the geometric mean exceedance may be the result of just one very dirty sample. Three additional stations in Section 1 exceeded Massachusetts state water quality standards for Enterococcus levels. Each of these stations is located in the upstream, fresh water portion of the section just downstream of Haverhill. New Hampshire bacteria standards for geometric mean calculations require at least three samples collected within a 60 day period. Since MRWC collected bacteria data only once per month in 2009, data frequency is insufficient for geometric means at the New Hampshire stations. In comparison to the sample data MRWC collected in 2007 and 2008, water quality in the Merrimack River seems to be generally improving during wet weather but diminishing, at least according to the more protective New Hampshire and CRWA standards, during dry weather. In general, the amount of bacteria in our samples has decreased. For example, the highest bacteria count MRWC collected in 2007 was 191,800 cfu/100 ml, but the highest collected in 2009 was only 1580 cfu/100 ml. The improvement during wet weather is probably the result of fewer combined sewer overflows (CSOs) throughout the river as cities such as Nashua, Lowell and Lawrence add stormwater treatment facilities, increase overall treatment capacity, and separate stormwater and septic sewer systems. The cause of the increase in dry weather bacteria amounts is currently unknown, but possibilities include an increasing number of failing septic systems and sewer pipes, more illicit connections whose discharges are no longer masked by CSOs, increased contamination from wildlife feces or a host of other potential causes. Additional data is required to determine if this trend is statistically significant. Appendix A summarizes the bacteria results from 2007 through

23 Mean Concentration (cfu/100ml) Tyngsborough MA State Limit = 126 cfu/100ml (E. coli ) Station MA State Limit = 35 cfu/100ml (Enterococcus ) Newburyport Figure 5. Geometric mean of E. coli and Enterococcus bacteria concentrations May through October 2009 in the Merrimack River. Stations with less than five samples, including all those in Section 5, were not included in the geometric mean calculation.

24 CSO ANALYSIS One of the goals of the MRWC s water quality monitoring is to begin determining sources of bacteria contamination to the Merrimack River. Because most of the samples collected over the past three years that exceed state water quality standards were collected during wet weather, project results indicate that most bacteria pollution in the river occurs during rain storm and snow melt events. Unfortunately, the effort to identify point versus non point sources of pollution is hampered by frequent combined sewer overflow (CSO) activity in the river, causing non point source concentrations to be obscured by CSOs and treatment plant bypasses in the five Merrimack River CSO communities of Manchester and Nashua, New Hampshire and Lowell, Lawrence and Haverhill, Massachusetts. While the Massachusetts CSO communities have been very helpful about providing data on CSO occurrences, the New Hampshire communities do not have this information available. The Lowell Regional Wastewater Utility (LRWWU), the Greater Lawrence Sanitary District (GLSD), and the Haverhill Wastewater Treatment Plant (HWTP) each reported CSO events during the monitoring season of May through October. Lawrence had the fewest with six days of CSO events during the monitoring season, Haverhill had 25 days (with one additional unconfirmed event on August 29), and Lowell had 28 days. For Lawrence and Lowell, these numbers were an improvement on the number of CSOs that occurred during the monitoring season in Between May and October 2008, GLSD reported 11 CSO days, HWTP reported 14 CSO days, and Lowell reported 33 CSO days. Figure 6 shows the Lowell daily rainfall and the days on which CSOs occurred in the Lowell area between May and October During this six month period, any local daily rainfall of more than three quarters of an inch caused a CSO in Lowell, and any daily rainfall amount over half an inch would have an 89 percent probability of causing an overflow (LRWWU 2010). While these numbers are disturbing to those using the river for swimming, boating and as a drinking water supply, they are a significant improvement over 2008 statistics. In 2008, a mere one third of an inch of rain was guaranteed to cause a CSO, and just a quarter inch would cause one 90 percent of the time. Figures 7 and 8 show similar rainfall and CSO information for GLSD in the Lawrence area and HWTP in Haverhill, respectively. The majority of events in Lawrence occurred when there was greater than one inch of rainfall in a day; however, as is evident from the June 12 rainfall, over an inch and a half of rain will not necessarily cause a CSO. In the case of GLSD, the intensity of rainfall and antecedent conditions are equally critical factors when attempting to predict overflows. The Haverhill sewer system appears shows similar characteristics to that of Lowell in that CSO events generally occurred when there was greater than one half inch of rain over the course of the prior day, and less rain was necessary to trigger CSOs when the system was recovering from previous rainfall. 23

25 Lowell Precipitation CSOs Total daily rainfall (inches) /1 5/15 5/29 6/12 6/26 7/10 7/24 8/7 8/21 9/4 9/18 10/2 10/16 10/30 Figure 6. Total daily precipitation in inches in Lowell, Massachusetts compared to days on which LRWWU CSO diversions occurred May through October, Date 2.5 Lawrence Precipitation CSOs 2.0 Total daily rainfall (inches) /1 5/15 5/29 6/12 6/26 7/10 7/24 8/7 8/21 9/4 9/18 10/2 10/16 10/30 Figure 7. Total daily precipitation in inches in Lawrence, Massachusetts compared to days on which GLSD CSO occurred May through October, Date 24

26 Haverhill Precipitation Confirmed CSOs Unconfirmed CSO Total daily rainfall (inches) /1 5/15 5/29 6/12 6/26 7/10 7/24 8/7 8/21 9/4 9/18 10/2 10/16 10/30 Figure 8. Total daily precipitation in inches in Haverhill, Massachusetts compared to days on which HWTF CSO occurred May through October, Date An example of the complications inherent in tracing bacteria sources in the presence of CSOs can be seen in the analysis of the highest bacteria concentration MRWC found in the Merrimack River in The sample was collected at station 19.1 on September 16 th, a dry weather day, near the mouth of the Little River in downtown Haverhill. The E. coli concentration of 1580 cfu/100 ml exceeded not only safe swimming standards, but also safe boating standards. Oddly, it was also the only sample collected in Section 2 on that day that contained high bacteria levels. Normally analysis would suggest the high bacteria concentration was due to a local source or illegal boat bilge dump and MRWC would schedule hotspot (areas of known or suspected high pollution) sampling in the area to determine if there was a chronic problem. However, a review of the CSO data indicates that GLSD experienced a CSO on September 12 th as a result of a brief but intense storm producing 1.5 inches of rain. Given the distance between the GLSD outfall and station 19.1 combined with the rates of flow during the four days between when the CSO occurred and the bacteria sample was collected, it is possible that the high bacteria counts were caused by the CSO. Under normal circumstances, a pollution plume will have dispersed sufficiently over the course of four days to have caused high bacteria counts at multiple monitoring stations, suggesting the source to be local to downtown Haverhill, but the presence of the Lawrence CSO cannot be fully discounted. As a result, MRWC is unable to determine the cause of the pollution, but will continue monitoring the location. 25

27 Temperature None of the water temperature readings collected in 2009 exceeded the Massachusetts state maximum temperature limit for Class B warm (28.3 C) or Class SB (29.4 C) waters. Frequent rain storms, especially during the first half of the 2009 monitoring season, resulted in higher average flows in 2009 (9,110 cu. ft./sec.) than during the same months in 2007 (5,251 cu. ft./sec.) or 2008 (8,479 cu. ft./sec.), reducing the amount of time water remained in the river exposed to sunlight and heating. Mean discharge May to October is approximately 5,448 cu. ft./sec. at the Merrimack River gauge in Lowell, Massachusetts, based on data collected from 1923 through 2009 (USGS undated). Water temperature in the Merrimack River followed the expected trend of heating during the warm summer months and cooling as the days shortened and the average air temperature cooled through the fall (Figure 9). The graph demonstrates the influence that warmer air temperatures have on the river, which is of concern as a result of the expected trend of temperature increases as a result of global warming. While the Merrimack is considered a warm water fishery, historically it has provided access to cold water streams and spawning habitat for anadromous fish and must remain cool enough for both the resident and transient fish and wildlife who depend on it for survival.. 26

28 Temperature (C) Air Temperature Water Temp. Section 1 Water Temp. Section 2 Water Temp. Section 3 Water Temp. Section 4 Water Temp. Section 5 0 5/1 5/21 6/10 6/30 7/20 8/9 8/29 9/18 10/8 10/28 Figure 9. Median daily water temperature in the Merrimack River between May and October 2009 combined with the average of the mean daily air temperature for the lower Merrimack Valley (Nashua, NH and Lowell, Lawrence, Haverhill and Groveland, MA). Date

29 Dissolved Oxygen No 2009 dissolved oxygen (DO) readings were under the water quality limit of 5 mg/l for warm water fisheries. In addition, Figure 10 shows that DO followed the expected seasonal trend of decreasing as the water increased and increasing again in the fall as water temperatures decreased. DO values collected on August 1, 2009 in Section 3 of the Merrimack River did not follow this trend, but this data was collected the day after a heavy storm, when increased river flow can add oxygen to the water. On August 26, 2009, dissolved oxygen readings throughout the monitoring trip in Section 1 were below normal for the Merrimack River. Many of the measurements were below 6.5 mg/l, and two taken at the Newburyport Wastewater Treatment Facility were slightly below 6.0 mg/l. Measurements of other water quality parameters that day were normal for fresh water at that time of year, though water temperatures that day were among the highest recorded during the 2009 monitoring season Dissolved Oxygen (mg/l) Water Quality Limit = 5 mg/l 2 Section 1 Section 2 Section 3 Section 4 Section 5 0 5/1 5/21 6/10 6/30 7/20 8/9 8/29 9/18 10/8 10/28 Figure 10. Median dissolved oxygen results by station in the Merrimack River between May and October Date 28

30 ph During the summer of 2009, MRWC water quality monitors measured unusually low ph values in the Merrimack River in Massachusetts. As apparent in Figure 11, ph values on four separate days, May 20 in Section 1, May 30 in Section 3, August 1 in Section 3, and August 19 in Section 4, were found to be well below the lower limit of 6.5 standard units set by the state for conditions acceptable in Class B waters. The ph of the Merrimack River water measured on these four days is equivalent to the ph of vinegar. The data has been reviewed for potential equipment malfunction and while it does not appear that the observed readings were in error, the possibility cannot be completely eliminated. On two occasions the low ph readings were confirmed using ph test strips. MRWC has consulted with MA DEP, EPA, U.S. Fish and Wildlife Service, and municipal water treatment plant officials regarding the ph results we recorded. None of the water quality experts contacted has been able to offer an explanation for the extremely low measurements found. On May 20, 2009, MRWC staff and volunteers monitored in Section 1 from Old North Canal in Haverhill to the Newburyport Wastewater Treatment Plant. At the Old North Canal site (station 14.1), ph values around 3.0 to 3.5 were observed and thought to be erroneous; therefore, they were not recorded. Calibration of the ph probe was checked and found to be within tolerance. Values observed at the next station downstream ranged from 3.3 to 3.5. From that point, ph values gradually increased at each station as the team headed downstream, rising from 5.2 at Cobbler Brook (station 10.6); to 7.0 at the Newburyport Wastewater Treatment Plant (station 2.7). During the return trip upstream, the team collected another ph reading as well as a sample of water at a point between stations 11.8 and The ph at this site was 5.8, and the water sample, evaluated with a volunteer s swimming pool test strips, confirmed the low ph readings. On May 30, 2009, MRWC staff and volunteers monitored in Section 3 from above the Essex Dam in Lawrence to Duck Island in Lowell. ph in this section of the Merrimack River ranged from 4.1 to 6.2 with the lowest values found at Fish Brook (station 33.4) and Trull Brook (station 36.3) and the highest values found at Bartlett Brook (station 32.2) and the Methuen Water Intake (station 31.4). Changes in ph from one site to another did not demonstrate any particular trend, unlike the data collected on May 20 th where ph consistently increased as the team moved downstream. On August 1, 2009, MRWC volunteers again monitored in Section 3 from above Essex Dam to Trull Brook. On this day ph ranged from 3.3 at five meters depth to 6.4 at the surface at the Methuen Water Intake (station 31.4). Low ph values were recorded at several other sites as well, but none demonstrated the extreme range of station On August 19, 2009, MRWC interns and volunteers monitored at the Lowell Motor Boat Club near station 41.1 and in the Pawtucket Canal. During this monitoring event, ph ranged from 3.2 in the Merrimack River at the Lowell Motor Boat Club to 6.5 at the Swamp Locks in the canal, though ph varied significantly by depth at most sites. 29